1. Volume 28, Issue 5, Pages 547-560 (March 2014)
MeCP2 Suppresses Nuclear MicroRNA Processing and Dendritic Growth by Regulating the DGCR8/Drosha Complex 
Tian-Lin Cheng, Zhizhi Wang, Qiuming Liao, Ying Zhu, Wen-Hao Zhou, Wenqing Xu, Zilong Qiu 
Developmental Cell 
Volume 28, Issue 5, Pages (March 2014)
DOI: /j.devcel
Copyright © 2014 Elsevier Inc. Terms and Conditions

2. Developmental Cell 2014 28, 547-560DOI: (10.1016/j.devcel.2014.01.032)
Copyright © 2014 Elsevier Inc. Terms and Conditions

3. Figure 1
Upregulation of Mature miRNAs in the Hippocampus of mecp2 Null Mice and the Role of MeCP2 in miRNA Processing
(A) Deep-sequencing data on the level of 720 mature miRNAs in hippocampal tissues from WT and mecp2 null (KO) mice. A total of 314 showed higher expression by ≥1.5-fold (red circles), and 63 showed ≤1.5-fold lower expression (blue circles) in KO mice, as compared to WT mice. The rest showed differences in expression level <1.5-fold (green circles). The pie chart depicts the percentage of three populations.
(B–D) Examination of primary, precursor, and mature miR-134, miR-383, miR-382, and miR-182 levels in MeCP2-shRNA-expressing neuron lysates. Lentivirus-harboring MeCP2 RNAi was seeded to mouse primary cortical neurons 2 DIV, and RNA samples were collected 5 days after viral infection. Primary, precursor, and mature miRNA levels were analyzed with qPCR. ctrl, control samples with GFP expression only.
(E) Examination of primary and precursor miR-134 levels in the presence of transcriptional blocker. Procedures similar to those described in (B) were used. Actinomycin D (1 μM) was applied for 3 hr prior to RNA collection. Primary and precursor miR-134 levels were analyzed with qPCR.
∗p < 0.05 (t test). See also Figure S1 and Table S1.
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2014 Elsevier Inc. Terms and Conditions

4. Figure 2 The Direct Interaction between MeCP2 and DGCR8
(A) Direct interaction between MeCP2 and DGCR8 shown by GST pull-down assay. The recombinant GST and GST-DGCR8 purified from bacterial sources were immobilized on glutathione beads, combined with purified His-MeCP2, washed, and analyzed on SDS-PAGE using antibodies indicated. Target bands are indicated by asterisks. WB, western blot.
(B) Interaction between various MeCP2 forms and DGCR8 in HEK293 cells. HA-tagged WT or mutated forms of MeCP2 were coexpressed with myc-tagged DGCR8. Cell lysates were immunoprecipitated with anti-HA antibody and analyzed on SDS-PAGE. MeCP2ΔN, N terminus (aa 1–91) deleted; MeCP2ΔMBD, methyl-DNA-binding domain (aa 92–161) deleted; MeCP2ΔN+ΔMBD, both N terminus and MBD (aa 1–161) deleted; MeCP2380, C-terminal segment (aa 380–492) deleted. IP, immunoprecipitation.
(C) Interaction between various DGCR8 forms and MeCP2 in 293T cells. Myc-tagged WT, N segment, or C segment of DGCR8 was coexpressed with HA-tagged MeCP2. Cell lysates were immunoprecipitated with anti-myc antibody and analyzed on SDS-PAGE. DGCR8N, (aa 484–773) deleted; DGCR8C, (aa 1–483) deleted.
(D) Direct interaction between MeCP2 and DGCR8C shown by GST pull-down. Procedures similar to those described in (A) were used.
(E) Direct interaction of MeCP2 with both RBD domains of DGCR8. DGCR8ΔR12, (aa 511–685) deleted; DGCR8ΔR2D, (aa 620–750) deleted.
(F) The role of RNA in the interaction of MeCP2 and DGCR8. HA-MeCP2 and myc-DGCR8 were cotransfected into 293T cells. RNase (final concentration of 0.25 μg/μl, 37°C for 30 min) was used to digest RNA prior to immunoprecipitation. Cell lysates were immunoprecipitated with IgG and anti-HA antibody and analyzed on SDS-PAGE. Lower panel shows quantitation of upper panel. Error bars are SEM.
(G) Schematic illustration of interaction domains of MeCP2 and DGCR8.
∗p < 0.05 (t test). See also Figures S2 and S3.
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2014 Elsevier Inc. Terms and Conditions

5. Figure 3
MeCP2 Interrupts the Interaction between DGCR8 and Drosha, Depending on Phosphorylation of Ser80
(A) Drosha binds to the C terminus of DGCR8. Myc-tagged WT, N segment, or C segment of DGCR8 was expressed in 293T cells. Cell lysates were immunoprecipitated with anti-myc antibody and analyzed on SDS-PAGE with Drosha western blotting.
(B) The interaction between Drosha and DGCR8 with the presence of WT MeCP2 and MeCP2380. Myc-tagged DGCR8 was coexpressed with HA-tagged WT MeCP2 or MeCP2380, respectively. Cell lysates were immunoprecipitated with anti-myc antibody and analyzed on SDS-PAGE. Lower panel shows quantification of (B).
(C) The interaction between Drosha and DGCR8 with the different amounts of MeCP2 protein. Different amounts of HA-tagged WT MeCP2 were coexpressed with myc-tagged DGCR8 as indicated. Cell lysates were immunoprecipitated with anti-myc antibody and analyzed on SDS-PAGE. Right panel shows quantification of (C).
(D) Phosphorylation of MeCP2 Ser80 is critical for binding with DGCR8. HA-tagged WT MeCP2, MeCP2S80A, or MeCP2S80D was coexpressed with myc-tagged DGCR8. Cell lysates were immunoprecipitated with anti-myc antibody and analyzed on SDS-PAGE with HA antibody. Lower panel shows quantification of (D).
(E) Western blot of MeCP2 Ser80 upon depolarization. Mouse primary cortical neurons were cultured 5–7 DIV and applied with the stimulus indicated. Samples were collected and analyzed with the antibodies indicated.
(F) Immunoprecipitation of endogenous MeCP2 and DGCR8 from lysates of mouse cultured cortical neurons. Mouse primary cortical neurons were cultured at 5 DIV and stimulated by 50 mM KCl for 30 min and immunoprecipitated with anti-MeCP2 antibody; samples were washed and analyzed on SDS-PAGE by western blot with the antibodies indicated. Lower panel shows quantification of (F).
(G) Interaction between Drosha and DGCR8 in neurons depleting MeCP2. Mouse primary cortical neurons were seeded with either GFP or MeCP2-shRNA-expressing lentivirus at 2 DIV and collected for coIP experiments at 7 DIV. Cell lysates were immunoprecipitated with either IgG or anti-Drosha antibodies and analyzed by SDS-PAGE with antibodies indicated. Lower panel shows quantification of (G).
(H) Interaction between Drosha and DGCR8 in brain lysates of mecp2 TG mice. Hippocampal lysates were collected from mecp2 TG and WT littermates and immunoprecipitated with either IgG or anti-Drosha antibodies and analyzed by SDS-PAGE with the antibodies indicated. Lower panel shows quantification of (H).
Error bars are SEM. ∗p < 0.05 (t test).
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2014 Elsevier Inc. Terms and Conditions

6. Figure 4
Intramolecular Interactions in MeCP2 Protein Regulated by Phosphorylation of Ser80
(A) FRET efficiency is increased in S80A mutant compared with the WT MeCP2 and MeCP2S80D. CFP and YFP are tagged onto N and C termini of WT MeCP2, MeCP2S80A, and MeCP2S80D. Fluorescence dot signals from ten cells of each condition were collected and measured. p < (Kolmogorov-Smirnov test).
(B) FRET efficiency is increased in MeCP2 FRET probe upon depolarization in neurons. CFP and YFP are tagged onto N and C termini of WT MeCP2 and are transfected into mouse primary cortical neurons. Under depolarization conditions, neurons were stimulated with 50 mM KCl for 3 hr. Fluorescence signals from ten cells of each condition were collected and measured. p = (Kolmogorov-Smirnov test).
(C) Direct interaction between GST-MeCP2-N (aa 1–305) and His-MeCP2-C (aa 306–492) shown by GST pull-down assay. The recombinant GST and GST-MeCP2-N purified from bacterial sources were immobilized on glutathione beads, combined with purified His-MeCP2-C, washed, and analyzed on SDS-PAGE.
(D) The interaction between MeCP2-N and MeCP2-C regulated by phosphorylation of Ser80. HA-MeCP2-N-S80A or HA-MeCP2-N-S80D was coexpressed with myc-MeCP2-C. Cell lysates were immunoprecipitated with anti-HA antibody and analyzed on SDS-PAGE. Lower panel shows quantification of (D). Error bars are SEM. ∗p < 0.05 (t test).
(E) In the unphosphorylated state (or the S80A mutant), the accessibility of the MeCP2 C-terminal domain (CTD) is restricted by a potential intramolecular interaction, between the MeCP2 CTD and the MeCP2 N-terminal region including S80. Phosphorylation of S80 increases the CTD accessibility, likely by disrupting this intramolecular inhibition of MeCP2 and forming an open conformation.
See also Figure S4.
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2014 Elsevier Inc. Terms and Conditions

7. Figure 5
Regulation of CREB, LIMK1, and Pumilio2 Expression by MeCP2 via Suppressing miR-134
(A) Examination of primary, precursor, and mature miR-134 levels in WT and MeCP2380 mutation forms of MeCP2-expressing neuron lysates. Lentivirus-harboring WT and 380 truncated forms of MeCP2 were seeded to mouse primary cortical neurons 2 DIV, and RNA samples were collected 5 days after viral infection. Primary, precursor, and mature miR-134 levels were analyzed with qPCR.
(B and C) Examination of primary, precursor, and mature miR-134 levels in WT, S80A, and S80D mutation forms of MeCP2-expressing neuron lysates. The experimental procedure is the same as that in (A).
(D and E) Effect of DGCR8 on MeCP2-regulated levels of miR-134 and CREB. MeCP2- and DGCR8-expressing plasmids were electroporated into mouse primary cortical neurons as indicated. RNA and protein samples were collected 5 days after electroporation. CREB levels were analyzed with western blots. Lower panel shows quantification of (D). (E) miR-134 levels were analyzed with qPCR.
(F) Regulation of CREB level by WT and MeCP2380 mutant. Mouse E15 cortical neurons were dissociated and electroporated with GFP, WT, and 380 truncated forms of MeCP2. Cells were lysated at 5–6 DIV and analyzed on SDS-PAGE.
(G) Regulation of CREB and LIMK1 protein level by MeCP2 overexpression. Mouse E15 cortical neurons were dissociated and electroporated with GFP, WT, and S80A mutant forms of MeCP2. Cells were lysated at 5–6 DIV and analyzed on SDS-PAGE.
(H) WT, but not S80A, MeCP2 is able to rescue CREB and LIMK1 protein levels after MeCP2 knockdown. Mouse E15 cortical neurons were dissociated and seeded with lentivirus-harboring shRNA or with shRNA-resistant rescue from WT or S80A form of MeCP2 for 5 days.
(I) miR-134 is required for MeCP2-regulated CREB and LIMK1 expression. Mouse cortical neurons were dissociated and electroporated with miRNA inhibitor control or miR134 inhibitor. Lentivirus-harboring GFP and shRNA against MeCP2 were seed at 2 DIV. Neurons were lysated at 7 DIV and analyzed on SDS-PAGE.
(J) Bidirectional regulation of miR-134 in mecp2 KO and TG mice. RNAs were collected from hippocampal lysates of mecp2 KO, TG mice along with their WT littermates. Levels of mature miR-134 were analyzed by qPCR. Error bars are SEM. ∗p < 0.05 (t test).
(K) Bidirectional regulation of CREB and LIMK1 protein levels in mecp2 KO and TG mice. Left panel shows that CREB and LIMK1 protein levels decreased in hippocampus of mecp2 KO mice. Hippocampal samples were collected from 2-week-old mecp2 KO mice and WT littermates. Right panel shows that CREB and LIMK1 protein levels were elevated in hippocampus of mecp2 TG mice. Hippocampal samples were collected from 2-week-old mecp2 TG and WT littermates.
(L) Bidirectional regulation of Pumilio2 protein levels in mecp2 KO and TG mice. Left panel shows that Pum2 protein levels decreased in hippocampus of mecp2 KO mice. Hippocampal samples were collected from 2-week-old WT and KO mice. Right panel shows that Pum2 protein levels were elevated in hippocampus of mecp2 TG mice. Hippocampal samples were collected from 2-week-old WT and mecp2 TG mice.
See also Figures S5 and S6.
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2014 Elsevier Inc. Terms and Conditions

8. Figure 6
MeCP2 Regulates Dendritic Growth by Inhibiting miR-134, and Dephosphorylation of MeCP2 S80 Is Required for Activity-Dependent Dendritic Growth
(A) Example picture of mouse primary cortical neurons transfected at 3 DIV with GFP alone and GFP with constructs indicated. Neurons were fixed and immunostained with GFP antibody at 6 DIV for measurement of dendritic length.
(B) Measurements of total dendritic length of each condition. ∗p < 0.01 (t test).
(C) Example picture of mouse primary cortical neurons with each construct transfected. Constructs expressing MeCP2 WT and mutations were transfected into cortical neuron 2 DIV and stimulated by KCl 48 hr after transfection. KCl with a final concentration of 25 mM was given to mouse primary cortical neurons for 36 hr before neurons were fixed and performed following experiments. Immunostaining with GFP antibody was used to examine the dendritic growth.
(D) Measurements of total dendritic length of each condition. ∗p < 0.01 (t test). A total 12–15 neurons from each condition were randomly selected and measured.
See also Figure S7.
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2014 Elsevier Inc. Terms and Conditions

9. Figure 7
MeCP2 Inhibits Dendritic Growth and Spine Formation in Hippocampal Slice Cultures
(A) Typical pictures of rat hippocampal slice culture neurons. Hippocampal slices from P7 rat were transfected at 3 DIV with GFP alone and GFP with MeCP2, MeCP2S80A, and MeCP2380 by gene gun delivery. Slices were fixed and immunostained with GFP antibody at 9 DIV.
(B) Measurements of total dendritic length of each condition. ∗p < 0.01 (t test). For each condition, 18 neurons were randomly selected and measured. At least 900 μm dendrites were analyzed for each condition.
(C) Typical pictures of spines of rat hippocampal slice culture neurons. Experimental procedures are the same as (A).
(D) Measurements of spine density over 10 μm dendrites of each condition. ∗p < 0.01 (t test).
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2014 Elsevier Inc. Terms and Conditions